Sources: OpenAI's ARR is nearing $70B, growing 70%+ since the beginning of Q3, with B2B revenue up 100%+; it added more consumer revenue in Q3 than all of 2025
Madison Mills /Axios:
Context & Ripple Effects
OpenAI's reported run-rate trajectory has accelerated from $3.4B in annualized revenue in 2024 to a $40B-plus estimate in August 2026. In April, OpenAI said enterprise supplied more than 40% of its $2B monthly revenue and was expected to approach consumer revenue by year-end.
The latest source-based figures suggest that enterprise adoption is advancing faster than that earlier mix forecast, while consumer subscriptions continue to add meaningful revenue. The story has been widely syndicated, underscoring investor and industry attention to revenue as a measure of frontier-model commercialization.
First-order effects
- OpenAI's reported B2B growth strengthens the business case for directing sales, product, and service capacity toward enterprise customers alongside its consumer products.
- A revenue run rate near the reported level gives OpenAI a larger commercial base from which to finance the compute-intensive delivery of its models.
Second-order effects
- Enterprise AI buyers gain a more financially substantial OpenAI vendor, raising the importance of product reliability, pricing, and commercial terms in their model-provider decisions.
- The reported pace of B2B growth puts greater pressure on other frontier-model providers to demonstrate enterprise revenue and distribution, not only model performance.
Third-order effects
- If enterprise revenue continues closing the gap with consumer revenue, frontier labs will increasingly be valued as recurring-revenue software and infrastructure businesses rather than primarily as research organizations.
- The revenue trajectory reinforces a market structure in which labs able to convert model usage into recurring sales have more capacity to sustain the large fixed costs of AI compute.
The trend: Frontier AI is moving from consumer-led adoption toward a commercial model where enterprise distribution and recurring revenue determine which labs can support expanding compute needs.